The Official (and only) Optical vs. Coaxial Thread

[Taxi]

This is the first thread on this subject. Ah, the good old days.

This thread probably has the most info, but it got ugly and we had to close it. (Edit - After re-reading this monstrosity, I noticed that most of what Justin posted in his long post there on page one had very VERY little to do with discussing digital cable specifics. Instead, it goes off on a tangent of analog wire [which everybody agrees effects audio signals noticably] and CD frequencies and word sizes. -Taxi)
Wash.
Rinse.
Repeat.

There are other threads in the forum about this. (A lot.) But these are the main ones.

[Istagi]

Optical and coax both suck. I use a 20Ghz wireless system and it works great, an no my hair loss has nothing to do with it.

[MrTony]

Thankfully, I have no choice. My player only has coaxial outputs.

[Enzian]

And remember; Don't look at the laser! You could burn your eyes out with that .25w LED!!

How many times have you heard that?!?!

Digital Coax is digital - but Optical is analog!! You see, light is actually a wave...

[Enzian]

I'm sorry - it was DTS that was analog.

[brian]

Good old jspartan...I actually had some civilized conversations with that guy

[AirSickMoth]

Here are some interesting Articles that I found.

http://www.audiovideo101.com/learn/a...nid=101&List=7

and

http://www.taralabs.com/industry_rev...ism_d2_htm.htm

[Mr. Furious]

The problem with many of the sources reporting "the true deal with Optical vs. Coax" is that they all have a vested interest in keeping the cable market alive. As long as there are people willing to pay thousands on cables, there will be a market to supply it. With the increased pressure to scientifically quantify the value of a digital interconnect, we're seeing more cable and HT sites (who often make ad revenue off of cable suppliers) making the case for expensive cables. However none of them are actually showing any real world statistics, scientific studies, or even raw data to support their argument. It's always an article based on theory (or limited / single user experience).

I'm willing to keep an open mind about this. I believe that there is a potential for "bit differences" between the different mediums. I believe that some error may occur due to a poor medium, but I also believe that any well designed digital system will have error correction. I'm not convinced that digital components cannot compensate for problems in the bit stream. I cannot believe that we're the only ones who have thought about buffering Audio/Video and resending the data when there is an error.

Further, evidence in other fields tells us that digital systems operate very consistantly from one application to another. AV is not a completely separate entity. That is not to say that there isn't any unique challenges - simply that some knowledge from one digital system can be applied to another.

Why has no one been willing to run a study on this? Could we come up with the conditions of a challenge / study?

[Wirehed]

I have the two largest audio/video stores in Sac willing to provide the space and products (various cables and speakers etc). I just have to set up the rest of it.

[1138]

Serioulsy Seamonkey, I'd be willing to come up for it. And any other people within a decent range are morethan welcome. Maybe we can set up a meet/test thing.

[Taxi]

I'll drive down, and take anybody who wants to go from the NW area. (Chromy/Thanos/chlngr1970/Morticia/Slade?)

[Mr. Furious]

What are the conditions of the test? How will certain things be insured?

Example: Volume settings, quality of equipment, random use of cables, studio group, double blind test conditions, etc.

What are we attempting to prove in this test - subjective listening tests or bit stream evidence from the decoders? Are you testing cheap coax vs. expensive coax?

I'm sorry I'm so far away - I'd love to be there!

[1138]

Strictly subjective. I'd work on settingup some thorough double blind tests. Not only satisfingly random but do eneough of everything too so that sample size isnt a problem (well, less of one I guess).

[Wirehed]

I'm thinking of using Audio F/X since the owner and a lot of the employees are really cool. Plus they don't care what people think. (as opposed to the other store that has expressed interest in letting me do this, they are an "official" supplier of Monster Cable and act, well, rather pushy).

http://www.audiofx.com/intro4.htm


I'll talk to the owner about how best to set this up. I'm thinking after closeing hours would work the best.

[Miggy]

I'm too far away as well, otherwise I'd be interested. I hpoe to see a detailed report afterwards!

Miggy, the Thief

[Wirehed]

I'm going to hire a writer to cover this "event". It will probably make a great story.

[tomdkat]

Hey 1138, I'm up in Hayward so you could pick me up on the way...

Peace......

[Reznor007]

The best way to do a test like this is to use the same equipment in every test, and only change the interconnect. You will want to use whatever is the best equipment at that particular store also.

Be sure to try everyting from the very cheap coaxial cable included with most components, very basic optical, and all the way up to the $10,000/3 foot coaxial cable in a blind test.

My own view on this is that is there were such huge differences in the digital signal, the decoder wouldn't be able to tell what was DTS/DD and what was PCM or anything else.

[Enzian]

Originally posted by Reznor007
My own view on this is that is there were such huge differences in the digital signal, the decoder wouldn't be able to tell what was DTS/DD and what was PCM or anything else.

Which is the conflicting idea behind this whole debate. If a digital signal is just a string of bits, to be error corrected using ome sort of OSI model; then what difference does it make? If there's a standard (say, DTS) then the player Tx DTS bitstream to the reciever... 1001010011001010010010010
is 1001010011001010010010010 no matter how it gets to the reciever. If, for some reason, the reciever Rx 1001010011101010010010010, it will be error corrected, I assume, on some level.

So, if 1001010011001010010010010 gets to the receiver via digital coax or if 1001010011001010010010010 gets to the reciever via an optical conx - what difference does it make?

And, if you swing by Boston on your way to the demonstration, pick me up!! I'd love to be disproven.

[1138]

I totally agree. It's not like with analog where minor changes affect in minor ways. It's been encoded, and if the information isn't right, it won't be decoded correctly and you get nothing (or something horribly wrong).

As an example (and this is overly simplified), what if I want to send 67. In binary, that's 1000011. Now, if the signal gets delivered wrong, who's to say how bad the outcome will be. If I read 1000001, then I get 65. Not too bad. If I read 11000011, then I get 99. Yowza. If I read 0000011, then I get 3. Oops. With a digital singal, minute changes cause both big and small changes.

[Wirehed]

Like I have siad before, that seems to be rational. However, obviosly (to me) there is something else at play here becuase all the cables sound different to me. And not by just a little either, I'm talking about a difference that should be easily observed.

[Taxi]

Yes, but differences can be "perceived" when there is an "expected" difference.

That is why it would be cool to do some double blind (or at least single blind) tests and see if you really can hear a difference.

The mind can play some amazing tricks on us..

[Tyler Durden]

Oh no! Not this old chestnut - The argument about digital being 1 and 0's is soooo old hat now and I have supplied vast amounts of reasearch previously to back up why. What I am interested in knowing is if 1138 and Taxi thinks that all DVD players and CD players sound exactly the same - the natural conclusion of their argument is that this is so. So my question is this - are you honestly saying that a $75 cd player will sound exactly the same as a $4000 cd player and are we all being conned? And will a $125 DVD player gives as good a performance - aurally - as a $4000 player? Surely if digital data is digital data (as you say) then they must!

[AirSickMoth]

Originally posted by 1138
I totally agree. It's not like with analog where minor changes affect in minor ways. It's been encoded, and if the information isn't right, it won't be decoded correctly and you get nothing (or something horribly wrong).

As an example (and this is overly simplified), what if I want to send 67. In binary, that's 1000011. Now, if the signal gets delivered wrong, who's to say how bad the outcome will be. If I read 1000001, then I get 65. Not too bad. If I read 11000011, then I get 99. Yowza. If I read 0000011, then I get 3. Oops. With a digital singal, minute changes cause both big and small changes.

Well, I'm not sure what kind of error checking a DD/DTS decoder has, but I assume it has it. The simple error checking on a computer does quite a bit (hehehe, "bit", is that a pun?). Any thing that is transmited in a computer is error checked on some level. If it didn't your computer would crash all the time (more than it does now for some of you).

When you send a string of bits, it comes with an error checking table. When you compare the bits in the table to the bits in the string, you can figure out if there was an error in the string. Using the error bits and redundancy bits, you can reconstruct the bit string like it was supposed to be. It is fairly simple and often done in hardware.

So, if some bits arrive before or after they should, the error checking should take care of it. Unless the error is very big.

Here is a small article dealing with Error checking on CDs

I'll try to look for better Error checkng stuff tomorow.

[Wirehed]

Originally posted by Taxi
Yes, but differences can be "perceived" when there is an "expected" difference.

That is why it would be cool to do some double blind (or at least single blind) tests and see if you really can hear a difference.

The mind can play some amazing tricks on us..

Yes, I agree with you here. BUT... one of the big differences with the different cables is the amount of base. The better the cable, the more bass I get.

One of the things I noticed with all the different cables I used was that the better cables would make the windows vibrate, and some cables would make the framed artwork in my house vibrate.

This isn't something I was just perceiving, it was actually happening. If I had been using a sound meter at the time, I'm sure there would have been a notable difference.

[Morticia]

Originally posted by Taxi
I'll drive down, and take anybody who wants to go from the NW area. (Chromy/Thanos/chlngr1970/Morticia/Slade?)

ROAD TRIP!!

twist my arm.....am I too late?

[1138]

Originally posted by Tyler Durden
The argument about digital being 1 and 0's is soooo old hat now and I have supplied vast amounts of reasearch previously to back up why.

Where? I'd like to see it. Really.

So my question is this - are you honestly saying that a $75 cd player will sound exactly the same as a $4000 cd player and are we all being conned? And will a $125 DVD player gives as good a performance - aurally - as a $4000 player? Surely if digital data is digital data (as you say) then they must!

If you're using a digital connection, and they both have the same bit bandwidth, then yes. Why not? Why use digital in the first place if it's no better than analog? Are we all being conned?!?!?!

So are you saying that a $25 CD-ROM drive will do a worse job of copying game files over to my computer than a $250 DVD-ROM/CD-R drive? Will I not have the game copied correctly?

I'd like somebody to explain to me just how these digital signals come out sounding different. Technical details with no limits. I can't honestly see how they could unless both Dolby and DTS have designed some of the worst digital compression schemes known to man. The fact that I either do or don't get my satelite signal from miles up in space to my house (with a working capability at about 60% correct bits) demonstrates the capability of digital encoding.